Classification of biomedical signal on IoT platform using support vector machine

The biomedical module of the paper consists of a Photoplethysmographic (PPG) signal and an ECG sensor. Due to the change of the amplitude of the optical sensor, the PPG sensor can sense the signal components with less power supply noise and electromagnetic interference. First of all, we observe the signal through the digital signal processor (DSP), then calculate the heart rate, blood oxygen saturation (SPO2) and blood pressure. The ESP8266 Wi-Fi module sends signal messages to the backend for classification. In this paper, we use Support Vector Machine(SVM) to classify these three feature vectors. The classification results will show the data into healthy, unhealthy, very unhealthy, and explore the accuracy of classification prediction by using SVM. Finally, the physiological parameters of the test object and classification result uploaded to the cloud storage and webpage display in order to provide the basis for big data analysis in future research.